SODA, a plug-and-play alignment framework that adopts a BPR-style contrastive objective to align recommender representations with target-side distributional representations against negative ones, is developed and demonstrated that SODA consistently strengthens diverse generative recommendation architectures.
Zi-Qiu Xue, Ding-Xian Wang, Yi-Meng Bai et al.· Proceedings of the 20th ACM...· 0 citations
Recent generative recommenders improve scalability by retrieving items through token generation instead of traditional ranking over large candidate sets. Yet their training signals are still dominated by discrete code prediction, which overlooks the soft assignment information naturally produced by the tokenizer. This...
Zi-Qiu Xue, Ding-Xian Wang, Yi-Meng Bai et al.· Proceedings of the 20th ACM...· 0 citations
GRASP is a rubric-aware on-policy self-distillation framework for LLM personalization that turns user-specific rubric aspects into fine-grained, token-level supervision, and Rubric-based Teacher Validation (RTV) is introduced, which retains only instances where the teacher sufficiently covers the target aspects, improv...
Yilun Qiu, Xiao-Yan Zhao, Cheng-Bing Wang et al.· 0 citations
Preference-Aligned Test-Time Training (P-TTT) is proposed, which explicitly encodes preference relations conveyed by contextual pairs into user-specific fast weights for personalized reward prediction and outperforms state-of-the-art methods by a large margin.
Bo-Hao Wang, Xiao-Yan Zhao, Yang Zhang et al.· 0 citations
It is argued that the core ideas underlying RAG are not new: foundational concepts such as integrating retrieval and language generation, knowledge augmentation, answer verification, and iterative query refinement had already been studied and instantiated in information retrieval (IR) and question answering (QA) resear...
Xiao-Yan Zhao, Yujie Cai, Yang Zhang et al.· 0 citations
Experiments on the LongLaMP dataset show that PrefReward outperforms non-personalized and retrieval-based baselines in both generation quality and personalization interpretability.
Yue Wu, Chengbing Wang, Yimeng Bai et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
